GA signaling protein LsRGL1 interacts with the abscisic acid signaling-related gene LsWRKY70 to affect the bolting of leaf lettuce

Abstract A variety of endogenous hormone signals, developmental cues, and environmental stressors can trigger and promote leaf lettuce bolting. One such factor is gibberellin (GA), which has been linked to bolting. However, the signaling pathways and the mechanisms that regulate the process have not been discussed in full detail. To clarify the potential role of GAs in leaf lettuce, significant enrichment of GA pathway genes was found by RNA-seq, among which the LsRGL1 gene was considered significant. Upon overexpression of LsRGL1, a noticeable inhibition of leaf lettuce bolting was observed, whereas its knockdown by RNA interference led to an increase in bolting. In situ hybridization analysis indicated a significant accumulation of LsRGL1 in the stem tip cells of overexpressing plants. Leaf lettuce plants stably expressing LsRGL1 were examined concerning differentially expressed genes through RNA-seq analysis, and the data indicated enhanced enrichment of these genes in the ‘plant hormone signal transduction’ and ‘phenylpropanoid biosynthesis’ pathways. Additionally, significant changes in LsWRKY70 gene expression were identified in COG (Clusters of Orthologous Groups) functional classification. The results of yeast one-hybrid, β-glucuronidase (GUS), and biolayer interferometry (BLI) experiments showed that LsRGL1 proteins directly bind to the LsWRKY70 promoter. Silencing LsWRKY70 by virus-induced gene silencing (VIGS) can delay bolting, regulate the expression of endogenous hormones, abscisic acid (ABA)-linked genes, and flowering genes, and improve the nutritional quality of leaf lettuce. These results strongly associate the positive regulation of bolting with LsWRKY70 by identifying its vital functions in the GA-mediated signaling pathway. The data obtained in this research are invaluable for further experiments concerning the development and growth of leaf lettuce.


Introduction
Leaf lettuce (Lactuca sativa L.), an annual vegetable crop in the Asteraceae family, is the most-eaten leafy vegetable globally and is highly valued due to its edible and medicinal properties [1]. Leaf lettuce is mainly cultivated in greenhouses or open fields, and during the process of cultivation is prone to high-temperature bolting (premature production of f lowering stems) [2,3]. High temperatures accelerate bolting and cause bitter leaves, reducing quality and production, thus resulting in limited leaf lettuce sales [4,5]. Hence, to increase yields as well as sales, the f lowering and bolting of leaf lettuce needs to be delayed while the cooking quality of the product needs to be maintained [6].
The f lowering of plants is a process that involves various pathways concerning hormone signaling molecules and f loral homeotic genes, whose combined effect causes a significant developmental change in the shoot apical meristem [7]. Many elements control the development of f lowers, including transcription factors, gibberellin (GA), jasmonic acid (JA), abscisic acid (ABA), salicylic acid, cytokinin, auxin (IAA), and ethylene [8]. For instance, increased GA sensitivity of Arabidopsis overexpressing ScNAC23 was detected in comparison with the wild type (WT), with significant acceleration of f lowering and senescence due to the presence of exogenous GA [9]. The f lowering period can be delayed as a result of the f loral inductive signal, the FLOWERING LOCUS T (FT) being transported to the shoot apical meristem, and suppressing the expression of SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) [10]. The interplay between FT and SOC1 is a critical regulatory mechanism that controls the timing of f lowering in plants. Therefore, several environmental variables, in conjunction with endogenous genetic elements, inf luence the bolting and f lowering of plants [11].
Among the plant hormones the GAs, a large group of tetracyclic diterpenes, perform a vital function in the development and growth of plants [12]. One of the major bioactive GAs, GA 3 , contains a C3-hydroxyl group and is generated from a basic diterpenoid carboxylic acid skeleton [13,14]. GAs are phytohormones that play prominent roles in controlling stem elongation and f loral induction [15]. Constitutive GA response mutants or GAinsensitive dwarfs have been employed to investigate the genes involved in GA signal transduction. These mutants enable the identification and characterization of genes that are linked to the transduction of the GA signal [16]. Arabidopsis GA-insensitive mutants, which can be either dominant or semi-dominant, are characterized by their reduced sensitivity to GA. In wheat, such mutants exhibit decreased height, while in maize the shortinternode D8 and D9 mutants, as well as mutants with the gainof-function gibberellin-insensitive (gai-1) allele, also display GAinsensitive phenotypes [16].
The agriculturally important GA response gene family DELLA is highly conserved [17]. The adoption of semi-dwarf plant types, many of which were subsequently shown to contain mutations in either GA homeostasis or DELLA proteins, was partly responsible for the increased cereal crop yields in the Green Revolution of the 1960s [18]. The DELLA proteins in Arabidopsis, such as REPRESSOR of ga1-3 (RGA), RGA-LIKE 1 (RGL1), RGL2, RGL3, and GA INSEN-SITIVE (GAI), perform various functions and redundantly inhibit the plant responses mediated by GA [12]. A family of putative transcription regulators, the GRAS family, has a domain subfamily known as DELLA (VHIID), whose members are encoded by RGA and GAI in Chinese cabbage [19]. Double null mutations, such as gai-t6 rga-24, in the ga1-3 background of Arabidopsis result in a 'wild-type' or GA 'overdose' phenotype, characterized by the increased promotion of GA-induced processes, including apical dominance, f lowering time, leaf expansion, and stem elongation [16]. However, a semi-dwarf phenotype with semi-dominant gainof-function mutation appears as a result of domain mutation of GAI and RGA [20]. RGL2 and RGL1 function as negative regulators, whereby the expression level of the former is temporarily elevated during the imbibition of the dormant seed, thus affecting germination, and the latter regulates various GA responses, such as the development of f lowers and the elongation of stems [16,21].
According to previous studies, some progress has been made in regulating stem extension and f lower development by GA signaling, but there are still many problems. Plant growth and development can be co-regulated by a variety of hormones (e.g. GA, IAA, JA, and ethene) and environmental factors (e.g. light) [22,23]. In Arabidopsis, ABA promotes DELLA gene translocation [24]. DELLA can bind to JAZ1, a jasmonate pathway repressor, and subsequently affect jasmonic accumulation [25]. JA is involved in the regulation of stamen growth and development by inf luencing the downstream elements of the DELLA protein, regulating its stability and activity [26]. Studies have also shown that GA signaling can eliminate the inhibition caused by DELLA proteins by upregulating some specific genes in the salicylic acid, ethylene synthesis, and JA pathways, thereby affecting leaf senescence [27]. The plant hormones GAs and brassinosteroids (BRs) have been found to interact with each other through a few key components. These components include the transcription factors of BRs and members of the DELLA protein family. By interacting with these components, GA signaling can modulate BR responses, and vice versa [28]. DELLA protein is involved in the signal transduction of GA, IAA, ethylene, ABA, and jasmonate and is essential in the development and growth of the plant.
Regardless of the observations above, it is still largely unknown how the DELLA protein regulates bolting and f lowering. Further investigation of the process at the molecular level and signaling pathways is needed. In conclusion, this experiment was conducted to further understand the role of the LsRGL1 gene in the GA pathway and the mechanism involved in regulating the development and growth of leaf lettuce. The RNA-seq analysis of leaf lettuce plants that exhibited a stable expression of LsRGL1 depicted considerable enrichment of differentially expressed genes (DEGs) in the 'phenylpropanoid biosynthesis' and 'plant hormone signal transduction' pathways. Eight related transcription factors were identified in the COG (Clusters of Orthologous Groups) functional classification, among which the expression of the ABA signaling-associated LsWRKY70 transcription factor varied significantly. The results of yeast single-hybrid (Y1H) experiments and β-glucuronidase (GUS) and biolayer interferometry (BLI) assays showed that LsRGL1 could interact with the promoter region of LsWRKY70, which confirmed that the LsWRKY70 gene performs a vital function in leaf lettuce in terms of its development and growth.

Promotion of the development and growth of leaf lettuce plants by exogenous GA 3
After several field experiments, two cultivars (bolting-resistant cultivar S24 and bolting-sensitive cultivar S39) with a difference of 50 days in field high-temperature bolting time were selected [29]. Transcriptome analysis was conducted on the leaves of the two cultivars at the bolting stage. The results showed that the relative expression differences of GA signal transduction-related DELLA genes in the two cultivars were >3-fold, which suggested that GAsignal transduction-related factors may perform a crucial role in bolting and f lowering of leaf lettuce, which is worthy of further study.
To investigate the effect of GA on the growth and development of leaf lettuce plants, GA 3 at a concentration of 150 mg L-1 was sprayed during the growth period of leaf lettuce. After 7 days of GA 3 treatment, plant weight, plant height, blade length, and stem length of leaf lettuce plants were higher than those of WT ( Fig. 1A and B). The difference in stem length was the most significant, indicating that exogenous GA 3 increased bolting and accelerated the growth process of the leaf lettuce stem. After transcriptome analysis of S24 and S39 leaves during the bolting stage, eight genes with >3-fold relative expression differences were identified among 38 DEGs related to GA signal transduction (Supplementary Data Table S1). The association of the expression of the above-mentioned genes with the GA pathway was determined, the data indicating a strong link between the two (Fig. 1C). We designed primers and determined gene expression by qRT-PCR (Supplementary Data Table S2). The expression levels of LsGA20OX2, LsGA3OX1, LsGASA1, and LsGA20OX1 increased after 7 days of GA 3 treatment. However, LsRGL1, LsGID1B, LsPAT, and LsGAI decreased (Fig. 1D). The decrease in LsRGL1 was the most significant. These data implied that exogenous GA 3 hormone could enhance the development and growth of leaf lettuce (e.g. bolting) by increasing the expression of the genes linked to the process.

LsRGL1 regulates bolting of leaf lettuce
To explore the function of the LsRGL1 gene, Agrobacterium tumefaciens containing gene overexpression (OE) or interference vectors, OE-LsRGL1 and RNAi-LsRGL1, respectively, was transformed into leaf lettuce plants. The data obtained indicated that plant weight and height, leaf length, and stem length of stably transformed OE-LsRGL1 plants were considerably less than those of WT or RNA interference plants, whereas RNAi-LsRGL1 plants were considerably larger than WT or overexpression plants. The plant height and stem length of RNAi-LsRGL1 plants were twice those of OE-LsRGL1 plants ( Fig. 2A and B). The qRT-PCR results showed a considerably enhanced expression level of OE-LsRGL1, whereas RNAi-LsRGL1 was considerably reduced in comparison with the control (Fig. 2C). LsRGL1 was detected in the nucleus by subcellular localization (Supplementary Data Fig. S1). Additionally, in situ hybridization analysis showed that LsRGL1 was expressed at stem tips, and the OE-LsRGL1 signal was significantly enriched compared with RNAi-LsRGL1 and WT plants. In contrast, RNAi-LsRGL1 f lower bud differentiation was more apparent (Fig. 2D). Thus, LsRGL1 was identified as a negative regulator of bolting.

Transcriptome analysis of leaf lettuce under stable expression and identification of genes with differential expression
To explore the molecular mechanism of LsRGL1 stable transformation, extraction of total RNA from four biological replicates of OE-LsRGL1, RNAi-LsRGL1, and WT leaf lettuce was performed, and the data were utilized to generate cDNA libraries. We obtained 85.2 Gb of clean data (each sample at least 6.71 GB) by Illumina high-throughput RNA sequencing from 12 samples. A total of 31 447 unigenes were collected from public protein databases [KEGG (Kyoto Encyclopedia of Genes and Genomes), GO (Gene Ontology) and COG] and 27 838 unigenes were annotated. The total length of clean reads ranged from 45 058 344 to 50 573 638 among the different libraries, and the percentage of 30 bases was >94.65% after filtering out low-quality and rRNA reads (Supplementary Data Table S3).
In the Venn diagram, there were 843 DEGs in WT, OE, and RNAi (Supplementary Data Fig. S2), where 152 genes were consistent with the expression profiles of RNAi-LsRGL1 < WT < OE-LsRGL1, whereas 248 genes were consistent with the expression profiles of RNAi-LsRGL1 > WT > OE-LsRGL1, subject to further categorization according to their GO annotations into various functional categories. The numbers of DEGs involved in 'catalytic activity' (CA), 'molecular function' (MF), 'biological processes' (BPs), and 'metabolic processes' (MPs) were the largest in the three groups under study (Fig. 3A). An increased number of the DEGs in most of the treatments were from the BP and MF categories. Notably, in the three groups, enrichment of KEGG terms linked to 'phenylpropanoid biosynthesis' and 'plant hormone signal transduction' was indicated (Fig. 3B). The DEGs were filtered according to an expression level |log2(fold-change)| > 1 and false discovery rate (FDR) < .05 in each pairwise comparison. We selected eight genes related to bolting and f lowering out of 54 transcription factors with significant differences. At the same time, eight plant hormone signal transduction and response proteins, such as genes linked to ABA, NAC transcription factors, and ethylene-responsive genes, were annotated in COG (Fig. 4). The results of the study imply that LsRGL1 may play a crucial role in the bolting process of leaf lettuce by regulating the expression of hormone-related genes.

COG gene expression analysis
To verify the expression of hormone-related genes annotated from COG, the correlation analysis showed that the expression of these eight genes was closely correlated with bolting, with LsWRKY70 (XP_023749894.1) having the strongest correlation (Fig. 4A). Eight representative genes were analyzed by qRT-PCR during leaf  (Fig. 4B). The transcriptome analyses of qRT-PCR and RNA-seq results were shown to exhibit strong correlation and integrity by analyzing the relative gene expression of the selected LsWRKY70 (Fig. 4C). Therefore, the LsRGL1 gene may affect lettuce bolting by regulating the related genes.

Both in vivo and in vitro biochemical tests verified that LsRGL1 could bind proLsWRKY70
To explore the molecular mechanism of LsRGL1 regulating leaf lettuce bolting-related genes, a Y1H assay was utilized to examine whether LsRGL1 binds to proLsWRKY70. LsRGL1 and proLsWRKY70 were ligated into pGADT7 and the pHIS2 vector, respectively, for self-activation and subsequent detection. The results indicated an absence of HIS3 nutritional reporter gene expression in control or even in cells that contained other constructs, whereas yeast cells that simultaneously contained both the aforementioned constructs (pGADT7-LsRGL1 and pHIS2-proLsWRKY70) expressed the reporter gene (Fig. 5A). Therefore, we deduced that this gene represents likely LsRGL1 targets. Nicotiana benthamiana leaves were utilized for the transient transactivation assay, which was the next step in the investigation into the transcriptional control of LsWRKY70 by LsRGL1. The coding sequence of LsRGL1 was inserted into a construct that contained the 35S promoter of the caulif lower mosaic virus. This construct was cotransformed into N. benthamiana leaves alongside another construct that contained GUS protein fused with LsWRKY70 promoters. An increase in the relative expression levels and staining of GUS was detected during the coexpression of LsRGL1 with proLsWRKY70 (Fig. 5B). Finally, the potential binding was examined further through BLI. Using the GCC-boxes of proLsWRKY70, probes labeled with biotin were designed and were subsequently divided into a, b, c, and d segments. The results indicate the binding of each to LsRGL1 (Fig. 5C). The resulting kinetic values obtained from the BLI assay, which was utilized to measure the binding affinities of LsRGL1 to proLsWRKY70, suggested considerably strong interactions. Hence, it was confirmed that LsWRKY70 is a direct target of LsRGL1.
The transcription level of genes is regulated by cis-acting elements in the upstream promoter sequence, while the induction conditions of proLsRGL1 and proLsWRKY70 are still unclear. To elucidate the transcriptional regulation mechanism of proLsRGL1 and proLsWRKY70, the transcriptional regulation mechanism sequence for cis-acting elements was determined using Plant-CARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/ html/). The resulting data showed that proLsRGL1 included GA corresponding elements and elements responsive to light, ABA, auxin-, and methyl jasmonate (MeJA), as well as other factors (Supplementary Data Table S4).

LsWRKY70 silencing by virus-induced gene silencing delayed bolting and improved lettuce nutritional quality
The role of the LsWRKY70 genes was examined further by suppressing their expression in S39 using VIGS in the TRV vector. The research analyzed below confirms that LsWRKY70 silencing can delay bolting while improving the nutritional quality of lettuce. Seven days after infection, leaf lettuce plants infected with the virus-carrying TRV2-LsWRKY70 showed a delayed bolting phenotype (Fig. 6A). qRT-PCR was employed for comparison and analysis of the gene transcription levels between TRV2-LsWRKY70 plants and control plants. Regarding the expression levels of LsWRKY70, the data of three independent analyses depicted a significant reduction in the silenced plants, whereas during examination of genes of the ABA signal transduction pathway the results showed that the transcription levels of ABSCISIC ACID-INSENSITIVE 4 (ABI4) and ABSCISIC ACID-INSENSITIVE MUTANT 5 (ABI5) were considerably higher than those in the control plants. At the same time, f lowering-related genes were detected, and the expression levels of FT, SOC1, and LEAFY (LFY) were decreased, while the expression levels of FLOWERING LOCUS C (FLC), and FLOWERING LOCUS M (FLM) were significantly upregulated compared with the control (Fig. 6B). Moreover, the levels of phytohormones were measured, showing that ABA and JA levels increased while GA 3 and IAA levels decreased (Fig. 6C). The results showed that the delay in bolting following silencing could affect the content of phytohormones. Therefore, the expression of phytohormones and ABA-related genes in TRV2-LsWRKY70 plants is important for lettuce bolting.
To verify the effect of TRV2-LsWRKY70 on the nutritional quality of leaf lettuce, physiological data were analyzed. The leaf lettuce plants expressing silenced TRV2-LsWRKY70 were significantly shorter than control plants, and plant weight, plant height, leaf length, and stem length were significantly smaller than WT or TRV2. When LsWRKY70 was silenced, represents RNAi-LsRGL1. The three replicates' standard errors of the mean are indicated by error bars. One-way analysis of variance (ANOVA) followed by Tukey's multiple range test was used to determine values with significant variation (P < .05), which are indicated by different letters placed above the bars. the contents of soluble sugar, malondialdehyde (MDA) and vitamin C (VC) of leaf lettuce were considerably reduced compared with the control. However, the cellulose content of leaf lettuce increased. The antioxidant enzyme activity assay conducted in leaf lettuce TRV2-LsWRKY70 and control showed that phenylalanine ammonia lyase (PAL), catalase (CAT), and superoxide dismutase (SOD) activities increased steadily in both. There was no major variation between WT and TRV2, while a rapid reduction in peroxidase (POD) activity was detected (Supplementary Data Fig. S4). These results imply that TRV2-LsWRKY70 could delay bolting and improve the nutritional quality of lettuce by regulating hormones and ABA-related genes.

LsRGL1 mediates gibberellin signaling during leaf lettuce bolting
Acting as a vital phytohormone, GA controls the regulation of various plant development processes [30]. The inhibitor of GA  respectively. proLsWRKY70-a, proLsWRKY70-b, proLsWRKY70-c, and proLsWRKY70-d all bound to LsRGL1 protein, and proLsWRKY70-a, proLsWRKY70-b, and proLsWRKY70-d were concentration-gradient-dependent. However, the promoter proLsWRKY70-c showed no significant concentration gradient dependence.
biosynthesis, uniconazole, causes yield increase in caulif lower and purple kale by suppressing bolting and f lowering, whereas these two processes are promoted in Chinese cabbage, mustard, and radish through the functions of GA [15,19,31]. Researchers have documented that the f lowering and bolting of leaf lettuce are inf luenced through DELLA proteins by GA, although the process has not been elucidated on the molecular level yet. In this study, GA 3 at a concentration of 150 mg L-1 was sprayed during the growing period of leaf lettuce to induce rapid bolting, and the resulting data were congruent with prior research (Fig. 1). Monitoring their expression and correlation, the link between DELLA genes, f lower-related genes, and cell elongation was investigated. Studies have shown that the expression of the DELLA genes is downregulated after a 3-hour GA treatment at the three-leaf stage of Chinese cabbage, indicating that these genes may inhibit stem growth, and degradation of this gene with GA 3 can initiate stem elongation and development [19]. All RGL1-transgenic plants in Arabidopsis showed delayed bolting, consistent with a lack of GA biosynthesis or GA perception [16]. Studies have shown that the overexpression of RGL1 or functional acquisition of GAI can partially save the early f lowering phenotype of WRKY75-overexpressing plants in Arabidopsis [21]. During stamen development in pumpkin f lowers, the transcription levels of GA receptor GID1b and DELLA inhibitor GAIPb increased significantly in mature f lowers [32]. Therefore, we speculated that LsRGL1 (a member of the DELLA family) inhibits stem growth and development, while GA 3 treatment degrades it and initiates stem elongation and development. These results support the view that DELLA proteins act as inhibitors of GA-responsive plant growth ( Fig. 2A). LsRGL1 negatively regulates the GA response and plays an important role in controlling stem elongation [12,16].

WRKY70 mediated regulation and crosstalk of hormones
The study of WRKY transcription factors and hormone crosstalk has recently advanced substantially. Research has shown that three groups of WRKY genes, OsWRKY24, OsWRKY53, and OsWRKY70, have the features of typical WRKY transcription factors, and negatively regulate the transcription of ABA and GA signaling [21,33,34]. Moreover, studies have shown that the wrky75 mutation leads to delayed f lowering in Arabidopsis and significantly accelerates f lowering after overexpression, positively regulating f lowering in an FT-dependent manner [21]. SlWRKY23 is hypersensitive to ethylene, JA, and auxinmediated root growth; the f lowering time of the transgenic plants was shortened, and the plants showed more inf lorescence branching [35]. In Arabidopsis, prior research has documented the link between DELLA and all three groups of WRKY proteins to examine the inf luence of GA on f loral initiation [7,21]. The exogenous application of GA 3 inf luenced the f lowering time, whereby the wrky12 and wrky13 mutants all showed different effects. GA 3 sped up f lowering time in the WT, whereas in comparison with the WT the phenotypic analysis depicted delayed f lowering in the wrky12 mutant and early f lowering in the wrky13 mutant, implying that the effect of the wrky12 and wrky13 mutations could be regarded as a positive and negative factors, respectively, in f lowering-time regulation. The f lowering modulated by GA shows partial dependence on the wrky mutations [36]. Moreover, this study found that in LsRGL1 transgenic plants, KEGG analysis detected enrichment of genes related to 'phenylpropanoid biosynthesis' and 'plant hormone signal transduction' (Fig. 3B). The data of these analyses indicated the vital function of hormones concerning their impact on bolting. Additionally, it was hypothesized that LsWRKY70 could promote ABA synthesis and activate ABA-dependent signaling pathways.
A recent study showed that the bZIP transcription factor genes ABI5 and ABI4 play a negative role in ABA-mediated LsFLC transcription inhibition and are also key components of ABA signal transduction pathways [37]. The relative expression levels of LsABI4, LsABI5, and LsFLC showed that silencing LsWRKY70 enhanced the tolerance of transgenic lines to osmotic stress. Hence, we speculated that this was due to the interaction of LsWRKY70 with f lowering-related genes and ABA signaling pathways (Fig. 6B). In Arabidopsis, ABI5 was found during the screening of ABA-insensitive mutants, and ABA has a strong induction effect on ABI5 nutritional expression [38]. In addition, FLC was shown to be the target of ABI4 and ABI5; FLC expression was directly regulated by these two genes, thus precisely controlling f lower transformation in plants [39,40]. In Chinese cabbage, SOC1 overexpression accelerated early f lowering and stem elongation, while SOC1 knockout significantly delayed bolting and f lowering [40]. These findings imply that the tolerance of the transgenic lines may be improved via the ABA signaling pathway, thereby delaying bolting, and that these effects can be linked to the elevation of the expression levels of these genes by LsWRKY70. The data obtained in this study were consistent with previous studies, where it was reported that overexpression of OsWRKY72 disrupted signal crosstalk between ABA signals and auxin transport pathways [40].
Additionally, early f lowering and decreased apical dominance were demonstrated in OsWRKY72 transgenic Arabidopsis.

Effect of hormone crosstalk on bolting of leaf lettuce
Hormones play a vital function in all aspects of the development and growth of plants. In lettuce, for instance, stems exposed to the combined effect of GA and auxin increased in thickness in comparison with those treated by GA solely at the same stage of f lower bud differentiation. This increased thickness was associated with the presence of IAA, which promoted the elevation of GA 3 content, leading to further stimulation of stem elongation and growth [41]. Several hormonal cues, including cytokinin and GA, control the start and progression of the stalkforming process in Chinese cabbage [30]. This study found that GA affected lettuce bolting and f lowering by inf luencing ABA. The data suggested that exogenous hormone administration can control the breeding cycle and offer more effective breeding methods. Therefore, it is essential to further elucidate how exogenous hormones can regulate the mechanism of lettuce bolting.

WRKY70 silencing treatment affects physiological changes in leaf lettuce
Through long-term co-evolution and ongoing natural selection, plants have gradually modified their survival strategies to cope with harsh climatic circumstances. This adaptation includes alterations in numerous physiological and metabolic pathways, such as the removal of reactive oxygen species (ROS), an increase in the synthesis of protein-protective complex, a decrease in osmotic regulators, variation in carbohydrate metabolism, and increased energy and lipid metabolism [42][43][44]. Moreover, the accumulation of soluble proteins, soluble sugars, and vitamin C can promote osmotic potential and play an important metabolic role in stress tolerance [44][45][46]. Soluble proteins help to safeguard enzyme function, stabilize protein synthesis, and control cytoplasmic acidity and alkalinity [47]. Soluble sugars, as binding substances of membrane lipids, play a critical role in maintaining membrane stability, and vitamin C is a major contributor to antioxidant capacity [47,48]. The findings of this study are congruent with the findings of earlier publications ( Supplementary  Fig. S4B). Following LsWRKY70 silencing treatment, soluble protein and sugar, as well as vitamin C content, increased to improve the stability and protect the integrity of the cell membrane of leaf lettuce by increasing the content of osmoregulatory substances.
Oxidative stress is caused by environmental stress, which interferes with cells' normal metabolism, causing ROS to accumulate through an imbalance of the formation and clearance of ROS [49]. The antioxidant enzymes SOD, CAT, and POD are produced under stress situations to scavenge harmful ROS, thus resulting in increased tolerance to these stressors [41,50,51]. The data of this research is congruent with prior studies (Supplementary Fig. S4C).
These results indicate that an extra response mechanism to stress was established by LsWRKY70 silencing-related treatment during leaf lettuce bolting and the enzymatic system was activated, thereby further enhancing the stress-adaptive ability, to improve the quality of leaf lettuce.
This research has identified the molecular pathways via which LsWRKY70 controls bolting. These findings show that LsWRKY70 interacts with LsRGL1 protein through ABA hormone-mediated bolting in leaf lettuce as a novel member of the bolting regulatory network. Further research into the regulation of LsWRKY70 transcription and translation during the switch from vegetative to reproductive growth will be intriguing. The conservation of this mechanism for other bolting-associated WRKY transcription factors is also a prospective area of research. This research has indicated the regulation of the onset and progression of bolting by LsWRKY70 as a novel member of the ABA-mediated signaling pathway.

Plant materials and growing conditions
The bolting-sensitive L. sativa cultivar S39, provided by Cathay Green Seeds (Beijing) Co., Ltd, was planted in the greenhouse of the Beijing University of Agriculture Experimental Station. Approximately 200 disease-free seeds were selected and sown in 50-hole disks. The growth conditions of lettuce were as published previously [5]. When the seedlings had grown to four leaves and one heart they were transferred to a 9-cm pot.

Exogenous GA 3 treatment
Following the ripening of the lettuce seedlings, a concentration of 150 mg L-1 GA 3 solution once every 3 days was sprayed onto them for a total of two times.

Plant transformation
Lettuce S39 plants were stably transformed by Agrobacteriummediated transformation using vectors containing overexpression and interference genes, OE-LsRGL1 and RNAi-LsRGL1. First, plump seeds were selected, sterilized, and grown on the medium for 3 days. The resulting leaves were cut with a blade, placed in a medium, and grown for 2 days in dark conditions. Next, the leaves were soaked in Agrobacterium solution for 15 minutes, followed by 2 days of growth in dark conditions. The extremely important process of transferring the leaves t o the medium for screening for resistance until new shoots emerged was carried out carefully. Finally, the shoots were placed in the rooting medium and transplanted when the roots were strong. The same parts of leaves from lettuce were extracted and wrapped in tinfoil. Primers from those listed in Supplementary Data Table S2 were used to screen the transformed plants.

In situ hybridization
Fresh specimens were fixed with RNA-free FAA fixative for 24 hours, routinely dewaxed, and washed. Anti-peeling slides were utilized, and the cover slides of cell crawling slides were treated with polylysine, fixed for 20-30 minutes, fully washed with diethy l pyrocarbonate (DEPC) water, dried naturally, and frozen at −20 • C for >2 weeks. The wet box comprised 5 × SSC (pH 7.5) (35 mL) + formamide (35 ml). Instructions for the chromogenic in situ hybridization (CISH) in situ hybridization kit were referred to for specific steps.

Yeast one-hybrid assay
The LsWRKY70 promoter sequence was cloned into the pHIS2 vector at polyclonal sites (EcoRI and BamH1) in front of the HIS3 reporter gene. The corresponding LsRGL1 transcription factors were cloned on the pGADT7 vectors (EcoRI and BamH1). A specific concentration of 3-amino-1,2,4-triazole (3-AT) was added to inhibit the background expression of this reporter gene, and suppressed the growth of false-positive bacteria. The Y1H technique was conducted as reported previously [53] (Supplementary Data Tables S1 and S2).

β-Glucuronidase staining assays for transient expression in leaf lettuce
The LsWRKY70 promoter fragment (1793 bp) was cloned into the pBI101 vector (BamHI and SmaI sites). At the same time, the LsRGL1 transcription factor sequence was cloned into the pBI121 vector (XbaI and SacI sites). The transient expression of these genes in lettuce was conducted using the Agrobacterium-mediated method, and GUS staining then determined their expression using prior research methods (Huayang Biology, China) [54].

Biolayer interferometry assay
BLI assays were conducted using an octet RED 96 system (Forte Bio, USA) with His-Tag sensors. The LsRGL1-His and GST-pGADT7 solutions were diluted to 20 μg ml −1 at a pH value of 4.5 using 10 mM sodium acetate, and the dilutions were employed to soak the sensors for 10 minutes. The dilution of the purified proL-sWRKY70 samples in BLI buffer to different concentrations was carried out to serve as analytes [54], according to the directions of the manufacturer. BLI was used to determine the dissociation constants (K D ) and the on-and-off rate for HIS-LsRGL1 binding to GST-LsWRKY70.

Construction of and infection with LsWRKY70 virus-induced gene silencing vectors
The LsWRKY70 fragment was inserted into the TRV2 vector, creating a 291-bp insertion fragment. To design the primers for this insertion, specific DNA sequences recognized by the BmaH1 and EcoR1 restriction enzymes were incorporated as limiting sites (Supplementary Data Table S2). Following the successful connection of the insertion fragment with the TRV2 vector, the resulting construct was transferred to Agrobacterium strain GV3101 to facilitate its introduction into the cells. The experiment was divided into a blank control group (WT), a negative control group (TRV2), and an experimental group (TRV2-LsWRKY70) according to previously published methods [5].

Malondialdehyde content and soluble sugar concentration
Soluble sugar content The soluble sugar test kit (Thermo Fisher Technology Co., Ltd., China) was utilized to perform the test per the instructions of the manufacturer. Then, 2 mL 10% (w/v) trichloroacetic acid (TCA) was added to 0.5 g of fresh lettuce lea ves, ground into a uniform mixture, and centrifuged. Afterward, 0.6% (w/v) thiobarbituric acid (TB A) was added to the supernatant liquid (volume 2ml), followed by 15 min in a 95 • C water bath, pla ced on top of the ice, and centrifuged. The absorbance values of the supernatant at 450 nm were det ermined. The quantitative formula is as follows: soluble sugar concentration (mmol L -1 )= 11.71 × A 450 .

Vitamin C (VC) content
The VC assay kit (Nanjing Jiancheng Institute of Biological Engineering, China) was utilized for testing according to the directions of the manufacturer. R1 (0.45 ml) was added to 0.15 g fresh lettuce leaves. The mixture was ground evenly and centrifuged. The supernatant was separated, followed by the addition of R2, R3, and R4. The mixture was vortically mixed and subjected to incubation at 37 • C for 30 minutes. Finally, the supernatant was isolated. Absorbance at 536 nm was measured, and the quantitative formula used for VC content (μg ml −1 ) was (measured OD value − blank OD value)/(standard OD value − blank OD value) × 6 × 4.

Cellulose content
The determination of cellulose content was performed using the Cellulose Content Detection Kit (Solarbio, China), following the manufacturer's instructions. To be precise, 0.1 g of lettuce leaves was utilized, and the test was carried out according to the protocol. Subsequently, the spectrophotometer was calibrated to a wavelength of 620 nm, and distilled water was used to adjust it to 0 nm.

Soluble protein content
The experiment was conducted according to the instructions of a soluble protein detection kit (Thermo Fisher Technology Co., Ltd, China). Initially, a standard curve was drawn with protein content as the abscissa and absorbance as the ordinate. Then, 5 ml buffer was added to 0.5 g fresh lettuce leaves. The mixture was ground to a uniform consistency and centrifuged. Next, 0.3 ml of supernatant was introduced into the test tube, followed by the addition of 5 ml of Coomassie Brilliant Blue reagent and appropriate shaking. Following a 2-minute rest period, the color at 595 nm was compared, the absorbance was calculated, and the protein content was checked using the standard curve. Finally, the following equation was used to calculate protein content: soluble protein content = C × VT/VS × WF × 1000 (mg/g).

Enzyme-linked immunoassay
In this experiment, an enzyme-linked immunoassay (ELISA) kit (Thermo Fisher Technology Co., Ltd, China) was used for detection and analysis [6]. Lettuce leaves (0.1 g) were ground into a powder with liquid nitrogen following the directions of the kit. The absorbance of GA 3 , ABA, IAA, and JA at 492, 490, 490, and 450 nm was determined, with the optical densities increased by 0.01 unit.